Problems with changing spell targets

Toy Sat 18 May 2002 01:43 PM #0

Just need a bit of help from anyone who can. :)

I wish to change the targets of some of my spells so you can cast them without needing a target- ie, fireball should be an area spell and not need a target. So, I changed the target on the fireball spell to ignore, and the mud crashed leaving me with this message:

Segmentation fault (CORE DUMPED)

Anyone know how to fix this so I can make my spells work right?

Thanx!
-Alexander

Nick Gammon Australia Forum Administrator Sun 19 May 2002 02:32 AM #1

Did you change the code itself? Can you post the changed function, or at least what you changed? What version of SMAUG did you use?

Toy Sun 19 May 2002 02:31 PM #2

I'm using smaug1.4a_mxp.tgz and I'm compiling with Cygwin. I'm still new to this coding stuff, and I didn't alter the code. I used the Area Editor program to change the target of fireball to ignore, then logged on and tried it. Dropped the core segment. I tried the same with acid breath, since I know that gas breath works on an ignore target. Changed acid breath's target and lost the core segment too. I attempted to change the code for lightning breath to work like gas breath in magic.c in an attempt to make them work alike:


ch_ret spell_gas_breath( int sn, int level, CHAR_DATA *ch, void *vo )
{
    CHAR_DATA *vch;
    CHAR_DATA *vch_next;
    int dam;
    int hpch;
    bool ch_died;

    ch_died = FALSE;

    if ( IS_SET( ch->in_room->room_flags, ROOM_SAFE ) )
    {
	set_char_color( AT_MAGIC, ch );
	send_to_char( "You fail to breathe.\n\r", ch );
	return rNONE;
    }

    for ( vch = ch->in_room->first_person; vch; vch = vch_next )
    {
	vch_next = vch->next_in_room;
        if ( !IS_NPC( vch ) && xIS_SET( vch->act, PLR_WIZINVIS ) 
             && vch->pcdata->wizinvis >= LEVEL_IMMORTAL )
          continue;

	if ( IS_NPC(ch) ? !IS_NPC(vch) : IS_NPC(vch) )
	{
	    hpch = UMAX( 10, ch->hit );
	    dam  = number_range( hpch/16+1, hpch/8 );
	    if ( saves_breath( level, vch ) )
		dam /= 2;
	    if ( damage( ch, vch, dam, sn ) == rCHAR_DIED || char_died(ch) )
	      ch_died = TRUE;
	}
    }
    if ( ch_died )
      return rCHAR_DIED;
    else
      return rNONE;
}
------------------------------------------------
ch_ret spell_lightning_breath( int sn, int level, CHAR_DATA *ch, void *vo )
{
    CHAR_DATA *victim = (CHAR_DATA *) vo;
    int dam;
    int hpch;

    hpch = UMAX( 10, ch->hit );
    dam = number_range( hpch/16+1, hpch/8 );
    if ( saves_breath( level, victim ) )
	dam /= 2;
    return damage( ch, victim, dam, sn );
}

changed to:
ch_ret spell_lightning_breath( int sn, int level, CHAR_DATA *ch, void *vo )
{
    CHAR_DATA *vch;
    CHAR_DATA *vch_next;
    int dam;
    int hpch;
    bool ch_died;

    ch_died = FALSE;

    if ( IS_SET( ch->in_room->room_flags, ROOM_SAFE ) )
    {
	set_char_color( AT_MAGIC, ch );
	send_to_char( "You fail to breathe.\n\r", ch );
	return rNONE;
    }

    for ( vch = ch->in_room->first_person; vch; vch = vch_next )
    {
	vch_next = vch->next_in_room;
        if ( !IS_NPC( vch ) && xIS_SET( vch->act, PLR_WIZINVIS ) 
             && vch->pcdata->wizinvis >= LEVEL_IMMORTAL )
          continue;

	if ( IS_NPC(ch) ? !IS_NPC(vch) : IS_NPC(vch) )
	{
	    hpch = UMAX( 10, ch->hit );
	    dam  = number_range( hpch/16+1, hpch/8 );
	    if ( saves_breath( level, vch ) )
		dam /= 2;
	    if ( damage( ch, vch, dam, sn ) == rCHAR_DIED || char_died(ch) )
	      ch_died = TRUE;
	}
    }
    if ( ch_died )
      return rCHAR_DIED;
    else
      return rNONE;
}

didn't quite work they way I wanted it to. I don't know if this is gonna be any help Nick, but it's the best I can do. :)
-Alexander

Amended by Nick to add forum codes for readability

Nick Gammon Australia Forum Administrator Mon 20 May 2002 01:29 AM #3

Hard to say for certain, but in your original spell_lightning_breath it had...

CHAR_DATA *victim = (CHAR_DATA *) vo;

Now with an "ignore" target, vo may well be NULL, in which case victim will be NULL, in which case saves_breath( level, victim) will cause a segmentation fault.